复杂系统行为预测的“机理+辨识”策略
首发时间:2006-09-29
摘要:复杂系统行为的预测是重要和困难的。由于机理模型预测只能对已知影响因子的已知作用方式预测,而辩识模型预测对已知影响因子的未知作用和未知影响因子的作用有一定的预测能力,因此我们提出复杂系统行为预测的“机理+辨识”模型策略:先进行机理模型的预测,之后再对实际记录数据与机理模型预测的差值进行辩识模型预测。
关键词: 复杂系统 机理模型 辨识模型 预测
For information in English, please click here
“Mechanism model + identification model” strategy for prediction of complex systems’ behavior
Abstract:Prediction of complex systems’ behavior is important and difficult. Mechanism models can only predict the results of known effect manner of known influencing factors, while identification model can predict in some extent the results of unknown effect manner of known influencing factors and the results of unknown influencing factors, thus here we suggest the “mechanism model + identification model” strategy for prediction of complex system’s behavior. In this strategy, prediction of mechanism models are first taken, then identification models are used to predict the difference between the actual historical data and results given by mechanism models.
Keywords: Complex System Mechanism Model Identification Model Prediction
基金:
论文图表:
引用
No.8565168601159505****
同行评议
共计0人参与
勘误表
复杂系统行为预测的“机理+辨识”策略
评论
全部评论0/1000